Search for: All records

The Saharan Air Layer (SAL) is a hot, dry, and dust‐laden feature that advects large concentrations of dust across the Atlantic annually to destination regions in the Americas and Caribbean. However, recent work has suggested the SAL may be a contributing factor to high‐impact drought in the Caribbean basin. While the SAL's characteristic dust loadings have been the focus of much previous research, fewer efforts have holistically engaged the co‐evolution of the dust plume, its associated convective environment, and resultant rainfall in Caribbean islands. This study employs a self‐organizing map (SOM) classification to identify the common trans‐Atlantic dust transport typologies associated with the SAL during June and July 1981–2020. Using the column‐integrated dust flux, termed integrated dust transport (IDT), from MERRA‐2 reanalysis as a SAL proxy, the SOM resolved two common patterns which resembled trans‐Atlantic SAL outbreaks. During these events, the convective environment associated with the SAL, as inferred by the Gálvez‐Davison Index, becomes less conducive to precipitation as the SAL migrates further away from the west African coast. Simultaneously, days with IDT patterns grouped to the SAL outbreak typologies demonstrate island‐wide negative precipitation anomalies in Puerto Rico. The SOM's most distinctive SAL outbreak pattern has experienced a statistically significant increase during the 40‐year study period, becoming roughly 10% more frequent over that time. These results are relevant for both climate scientists and water managers wishing to better anticipate Caribbean droughts on both the long and short terms. 

Abstract Long‐range aerosol transport is an important physical mechanism for ecological, biological, and hydrological elements of the earth system. Regarding the latter, regional climate models have no way of assimilating future aerosol concentrations, so dust aerosol emissions must be parameterized using local landscape and meteorological conditions. The purpose of this study is to evaluate the accuracy of different dust emission settings within the Weather Research and Forecasting model coupled with chemistry (WRF‐Chem) to facilitate future dynamical downscaling work. This study performs nine WRF‐Chem hindcasts, each utilizing a different dust emission configuration, from 1 March to 31 May 2015, coinciding with a Saharan air layer (SAL) dust outbreak during the 2015 Caribbean drought. WRF‐Chem aerosol optical depth (AOD) and Gálvez‐Davison Index (GDI), a convective forecasting parameter, are validated against analogous MODIS, AERONET, and ERA5 products. In aggregate, the GOCART dust emission scheme with Air Force Weather Agency modifications (GOCART‐AFWA) achieved the best balance between AOD and GDI accuracy when employing the default tuning constant (1.00). As the schemes emitted dust more aggressively, WRF‐Chem produced warming at 500 hPa, reducing GDI over the central and eastern Atlantic near the modeled dust trajectory. Though AOD was generally too low over the southwest Atlantic, the eastern Caribbean occupies a transition zone between negative and positive AOD biases where this field was hindcast with relative accuracy. Meanwhile, areas with positive AOD biases were associated with negative GDI biases (and vice versa) indicating the covariability between SAL dust loadings and thermodynamic conditions in the tropical north Atlantic.